>let me fail to compile your completely working source code after I upgrade to a new version
>>107713960C23 deprecated K&R functions and implicit int for types, not a compiler issue
How dumb are you? The released compiler compiles itself just fine, your code is just wrong.
>>107714122does this still compile?int main(argc, argv, envp)int argc;char *argv[],char *envp[]{return EXIT_SUCCESS;}
int main(argc, argv, envp)int argc;char *argv[],char *envp[]{return EXIT_SUCCESS;}
>>107714178It never compiled.
do i see the fun in GCC?
>>107714232Nigger I'm a white man, I write my own fucking compiler
>>107714228oops, you're right it should've been ; instead of ,warning: old-style function definition [-Wold-style-definition]int main(argc, argv, envp)int argc;char *argv[];char *envp[];{return 0;}
int main(argc, argv, envp)int argc;char *argv[];char *envp[];{return 0;}
>>107714178If you add semi colons, -ansi, and use a proper return value, yeah
>>107714271This compiles without -ansi, now omit the return type of main and it won't For some reason if you omit the type of arc it won't compile even in -ansi mode
>>107714140True unless you compile with -Werror -Wall -Wextra -Wconversionthey'll add new warning every now and then, and actually introduced a false positive in GCC9 where you ABSOLUTELY could not discard the return value of write(2)
>>107714271nigga what the duck am I looking at, and why did this compile at any point in time?
>>107714335it still compiles https://godbolt.org/z/exTzPP9Wc
>>107714335it just does ok
>>107714335int main(argc, argv, envp) int argc; char *argv[]; char *envp[];{ 3[argv]; return 0;}>warning: statement with no effect [-Wunused-value]> 6 | 3[argv];
int main(argc, argv, envp) int argc; char *argv[]; char *envp[];{ 3[argv]; return 0;}
>>107714403>3[argv];Will be deprecated in c2y
>>107714256You've never even seen a line of code in your life.
>>107714335C used to look a lot different. It took a while for the modern syntax to emerge. Back when C was actually developed and got new features.Now all the good features are compiler extensions because the retards in charge of the C standard somehow think that not implementing features people want and use is a good thing.Pretty much all big C projects rely on compiler specific language extensions. C with extensions is a pretty good language, too bad it's not portable. Someone is going to respond to this saying that nobody needs those extensions. Yeah, sure. Clearly all the people working on Linux are dumber than you, random /g/ poster who never made anything of value.
>>107714493Retard C got 2 substantial updates since C99.
>>107714508They added nothing of value.
>>107714513>threads>checked arithmetic>function attributes>nothing of valueSpoken like a true nocoder.
>>107714567Those are nice, but they don't solve the real issues the language has.Standard library threads are nice, but cross platform threading libraries weren't exactly rare before.Checked arithmetic is nice, but could easily be achieved on all compilers before. Nice to have, but nothing fundamental.And the same goes for function attributes. Most are just small compiler hints. [[nodiscard]] is the only one I regularly use. Can you tell me how they revolutionized the language for you?
>>107714614>"THEY ADD NOTHING SO PPL HAVETO RELY ON EXTENSIOORINOS SOOOOOOO HORRIBLE">language adds features so extensions aren't needed anymore>"THEY ADDED THINGS WE ALREADY HAVE EXTENSIOORINOS FOR SOOOOOOOOO HORRIBLE"Don't think people don't notice when you're this fucking disingenuous.
>>107714614In C2y we is getting counted() for size of arrays, declaring variables inside if statements, ranged case's and labeled breaks to break from two loops or switches inside loops, how is this not hot
>>107714637It's nice for them to standardize things he previously needed a billion #if COMPILER_A bullshit macros for, really. But how about things like defer {cleanup_block} to get rid of the countless memory and resource leaks and double frees that plague C code everywhere, anonymous functions, ADTs?
>>107714680Defer is in the process of being standardized, but when it's there guaranteed you'll find an excuse to keep bitching.
>>107715218Actually yes, that's based as fuck.
I NEED THIShttps://clang.llvm.org/docs/LanguageExtensions.html#vectors-and-extended-vectors
>>107714436Can you post the paper that deprecates this?
>>107715479The one about the provenance of pointers iirc
>>107715218>you'll find an excuse to keep bitching.I certainly will bitch about it because as usual they will somehow make all the wrong choices and the result will be barely usable, just like it happened with #embed
